thermal kill switch

Joel Jaeggli joelja at
Wed Oct 23 09:57:36 PDT 2002

On Tue, 22 Oct 2002, Andre Lehovich wrote:

> We had the air-conditioning fail yesterday.  Caught it in
> time to shut down by hand, but we won't be so lucky next
> time.  RGB's book recommends a thermal kill switch, but
> doesn't give details on implementation.  One obvious idea is
> to have a daemon monitor lm-sensors and shutdown each node
> as it gets too hot.  This is easy and cheap.

most server mainboards can be set to kill the machines off at a certain 
threshhold at the very minimum. shutting down gracefully is more 
desireable in terms of the state of your disk.
> But, is there anything better?  We have not yet had the
> electric and cooling contractors refit our server room.

having one fitted to your epo (Emergency power off) bus ought to be fairly 
straight forward. That ought to kill everything at once.

also, having your chillers/ac integrated into your monitoring system, goes 
a long way towards catching it in time.

>  Is
> there anything we should have them install during the
> rewiring?  What are the pros/cons of a room-wide kill switch
> vs. the lm-sensors approach?

the big problem is if you have it set to a high enough threshhold that 
you won't accidentaly set it off say 45 or 50c it won't shut off fast 
enough to prevent hardware damage.

> Thanks,
> --Andre
> _______________________________________________
> Beowulf mailing list, Beowulf at
> To change your subscription (digest mode or unsubscribe) visit

Joel Jaeggli	      Academic User Services   joelja at    
--    PGP Key Fingerprint: 1DE9 8FCA 51FB 4195 B42A 9C32 A30D 121E      --
  In Dr. Johnson's famous dictionary patriotism is defined as the last
  resort of the scoundrel.  With all due respect to an enlightened but
  inferior lexicographer I beg to submit that it is the first.
	   	            -- Ambrose Bierce, "The Devil's Dictionary"

More information about the Beowulf mailing list